The goal of this project is the identification of nonlethal mutations in the DEN virus genome which when incorporated into the infectious cDNA clone (project# Z01 BK 07001-03) will reduce virulence. (1) Membrane integration of the C protein. The flavivirus C contains a conserved internal 20-aa hydrophobic domain. We demonstrated that this domain functions as a signal-anchor (SA) to mediate the membrane integration of C. The unusual membrane orientation of C was demonstrated through the analysis of the membrane integration of selected mutants. Function of this SA is essential for efficient recognition of the prM signal peptide. (MS in preparation.) (2) Mode of association of the flavivirus prM and E proteins. PrM and E are integrated into ER membranes co translationally and form a detergent-stable heterodimer which remains intact until prM is cleaved to produce M in acidic post-Golgi vesicles prior to exocytosis of mature virus particles. The portion of prM that is discarded contains conserved negatively charged domains which may be required for heterodimer formation with E. Mutants have been constructed to test this hypothesis in vivo and in vitro in a cell-free translation system using microsomal membranes as a source of ER. (3) Effect of mutations in the DEN NS2B protein on protease activity of NS3. To further characterize the requirement of NS2B for the protease activity of the virus-coded protease, NS3, we mutagenized a 40-aa domain in NS2B indispensable for protease activity. >60 mutants were isolated, and autocleavage at the NS2B-NS3 site was analyzed in vivo and in vitro. Some mutations resulted in partial defects in cleavage, thus defining potential attenuating mutations. Work has been initiated to engineer some of these mutations into an infectious cDNA clone of DEN. (4) Mutagenesis of the NS3 protease. There are four domains in the N terminal portion of NS3 with sequence homology to serine proteases. We have begun a limited mutagenesis of some of these conserved residues in order to characterize their requirement for protease activity and in an effort to isolate mutants with partial defects in cleavage. (5) Cleavage of the NS1-NS2A site. This cleavage is known to occur after a conserved motif which is a consensus site for cleavage by the ER resident protease, signal peptidase (SP). However, the motif does not occur in the context of a signal peptide, a requirement for recognition of the motif by SP. We have demonstrated that cleavage takes place in vitro in the presence of microsomes containing ER and its resident proteases. ER depleted of soluble intraluminal enzymes was still active for cleavage, suggesting that the cleavage enzyme is ER membrane-bound. The hypothesis that the NS1-NS2A entity is by itself an autoprotease was investigated by selectively mutagenizing all aas that could possibly be required at the active site of known types of proteases. Results suggest that NS1-NS2A is not an autoprotease and that NS1-NS2A cleavage is catalyzed by SP in a novel fashion. (MS in press in J. Virol.)